Large data center and enterprise networks (e.g., service provider networks) following a two or three tier architecture (e.g., an access layer, an aggregation layer, and a core layer) typically have more than one network device (e.g., a chassis, a switch, a router, etc.) at the aggregation layer. Such an arrangement provides network device-level resiliency. Access layer network devices will typically have connections (e.g., links) to all the aggregation layer network devices for link-level and device-level resiliency. Furthermore, the aggregation layer network devices are interconnected by links which are in the forwarding state all the time. Therefore, if a protocol (e.g., a spanning tree protocol (STP) or other protocols) is executed between two aggregation layer network devices and one access layer network device, a loop may be formed between the three network devices. If failover for the protocol (e.g., the STP) is in the order of seconds, a multi-network device link aggregation may be implemented to prevent looping.
A link aggregation (e.g., as set forth in IEEE 802.3ad) is a computer networking term which describes using multiple links (e.g., Ethernet network cables and/or ports in parallel) as one logical port to increase the link speed beyond the limits of any one single link. Other terms used for link aggregation may include Ethernet trunking, port teaming, network interface card (NIC) bonding, link bundling, and/or a link aggregation group (LAG). A LAG will be used hereinafter to refer to link aggregation.
Current implementations of a multi-network device LAG are either restricted to two network devices (e.g., a two switch aggregation layer model) or have one of the links of the multi-network device LAG in a passive mode. The passive link will switch over to an active mode when an active link becomes disabled (e.g., non-operational). However, these arrangements provide no scalable way to implement a LAG across multiple (e.g., more than two) network devices.